CN107193639B - Multi-core parallel simulation engine system supporting combined combat - Google Patents
Multi-core parallel simulation engine system supporting combined combat Download PDFInfo
- Publication number
- CN107193639B CN107193639B CN201710417606.9A CN201710417606A CN107193639B CN 107193639 B CN107193639 B CN 107193639B CN 201710417606 A CN201710417606 A CN 201710417606A CN 107193639 B CN107193639 B CN 107193639B
- Authority
- CN
- China
- Prior art keywords
- model
- simulation
- scheduling
- management module
- entity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 168
- 238000000034 method Methods 0.000 claims abstract description 34
- 230000008569 process Effects 0.000 claims abstract description 18
- 230000003993 interaction Effects 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 7
- 238000012800 visualization Methods 0.000 claims description 5
- 230000003542 behavioural effect Effects 0.000 claims description 4
- 230000006854 communication Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 2
- 230000002452 interceptive effect Effects 0.000 claims description 2
- 238000007726 management method Methods 0.000 description 37
- 239000003795 chemical substances by application Substances 0.000 description 33
- 239000000306 component Substances 0.000 description 12
- 238000011161 development Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000003068 static effect Effects 0.000 description 6
- 230000015572 biosynthetic process Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 230000006378 damage Effects 0.000 description 5
- 238000005755 formation reaction Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 238000005094 computer simulation Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000005538 encapsulation Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000010391 action planning Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002567 autonomic effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-core parallel simulation engine system supporting combined operation, which solves the problem that the real-time performance is easily influenced when a traditional combined operation system adopts step length to advance logic time. The system comprises a scheduling model management module, a thread management module, an external interface management module and a high-level architecture management module. The system firstly allocates target nodes for the simulation entity, so that the total computation of the model on each node is equivalent; and then, generating a scheduling plan table of each node based on a load balancing principle through a scheduling model management module, distributing simulation step length for the model, adjusting the scheduling plan table in the simulation process, and adjusting the simulation step length of the damaged entity and the generated new entity. The invention can automatically divide the scheduling table according to the running period of the model and the system step length, allows the entity to adopt different physical models or behavior models for simulation according to the requirements, and supports the real-time scheduling of large-scale simulation and high-fidelity combat models.
Description
Technical Field
The invention belongs to the technical field of computer simulation, and relates to a multi-core parallel simulation engine system supporting joint operations.
Background
With the deep development of military revolution, the simulation system plays an important role in a plurality of military fields such as assistant decision, command training and the like for the combined combat simulation system of a plurality of military varieties. The information-based war becomes the main form of future war.
The traditionally developed combat simulation engine has the limitations such as no support for parameterized assembly of models, no support for reuse of models, slow simulation speed of models, no capability of providing general interface service for model developers, no support for flexible expansion of systems, and the like. Because the current requirements on the simulation scale and the fidelity of the multi-weapon combined operation are further improved, and the simulation of each virtual operation battlefield has a specific weapon distribution mode and operation environment, finding an efficient and real-time engine scheduling algorithm is very important and difficult for the simulation of the whole battlefield.
The scheduling problem of the simulation engine system is a decision problem consisting of various qualitative and quantitative factors, and is particularly applied to the situation of multi-weapon combined combat. The current simulation engine has a plurality of limitations in the field of application to multi-weapon combined combat, and researches related to real-time scheduling of the simulation engine comprise 3 aspects of operation support environment, performance improvement of application and real-time scheduling of models, wherein the real-time scheduling of the models plays a decisive role.
The real-time scheduling of the model is to reasonably arrange the execution of the model to meet the deadline requirement, and can be divided into static and dynamic types: the former means that an operation schedule is arranged before operation; the latter refers to dynamically determining the model that needs to be executed at runtime, e.g., the earliest deadline algorithm assigns the task with the earliest deadline the highest priority. The earliest deadline algorithm is the most commonly used real-time scheduling algorithm and is optimized for different applications. Recent studies have applied the earliest deadline algorithm directly to HLA (high level architecture) environments, but not to combat simulation applications, and performance has not been verified.
A new generation of joint combat employs assembly and reuse techniques to support rapid development of models, such as JSAF and One SAF from the US army. The method is mainly characterized in that a multi-resolution model is supported and verified, and by using a scenario meeting military scenario description language standard specification, three fields of weapon research and development, development and employment, exercise training and analysis can be simultaneously supported. However, at present, no simulation engine system which can be applied to combined operations of multiple weapons is mastered at home, and workers in the field are greatly limited in technology during simulated operations.
Disclosure of Invention
The invention provides a multi-core parallel simulation engine system supporting combined combat, which aims to solve the application problem of a simulation engine in a combined combat system, carries out multi-core distributed and parallel transformation on the traditional combat simulation engine system, and is mainly suitable for a military force model. The simulation engine system divides a scheduling table according to the model operation period and the system step length, allocates the simulation step length for the model based on the principle of load balance, and provides support for the effective operation of the engine system by adopting a dynamic and static combined scheduling algorithm.
The multi-core parallel simulation engine system supporting the joint combat provided by the invention runs on the nodes distributed by adopting the Ethernet. The system comprises a scheduling model management module, a thread management module, an external interface management module and a high-level architecture management module. The scheduling model management module manages and schedules the simulation process of the model, manages the model in a queue mode for the thread management module to call, and sends a signal to the thread management module when the model simulation settlement is finished. And the thread management module creates and dispatches threads, and multithread dispatching takes out the model from the model queue for simulation settlement. And the external interface management module is an interface for the scheduling model management module to interact with the outside. The high-level system structure management module provides a high-level system structure HLA integrated interface, performs external simulation interaction with the operation support environment, and realizes the collaborative simulation with member nodes.
The multi-core parallel simulation engine system firstly allocates target nodes for simulation entities, so that the total computation of models on each node is equivalent; then, a scheduling model management module generates a scheduling plan table of each node, a simulation step length is distributed to the model, a scheduling table is adjusted in the simulation process, and the simulation step lengths of the destroyed entity and the generated new entity are adjusted.
The step of allocating simulation step length for the model by the scheduling model management module comprises the following steps: aligning the running periods of the models; initializing the idle rate of a processor of each system step length in a scheduling period to be 1; and generating a scheduling schedule, wherein the scheduling schedule is organized according to different model operation cycles, and comprises model step sizes and the number of system step sizes contained in the model step sizes.
The scheduling model management module generates a scheduling schedule according to different operation periods T of the models in a sequence from small to large1,T2,…,TsAnd traversing, creating a scheduling table of each system step length in each running period, and finally forming a scheduling schedule. For operation period TiCalculating the operating period TiThe number of system steps contained in L-Ti/TstepI ═ 1,2, …, s; will have a run period TiAll the models form a queue according to the sequence of the execution time from large to small, traverse the model queue, and execute the following (1) to (3) to create the operation period TiScheduling Table of step length of each internal systemij,1≤j≤L；
(1) Slave operation period TiFor each system step size, selects the maximum idle among the processor idlenessm；
(2) The current model is allocated to the mth system step length and recorded in the TableimPerforming the following steps;
(3) updating the processor idle rate of the mth system step size and the subsequent affected system step size; the processor idle rate update of the m + L system step is: idlem+L*l-T/Tstep(ii) a Wherein l is more than or equal to 0<Ts/Ti；
And when the idle rate of the processor of the system step size is less than 0, stopping distributing the model for the system step size.
The invention has the advantages and beneficial effects that:
1) the multi-core parallel simulation engine system supporting the combined operation aims at real-time scheduling of a simulation engine of a combined operation model and supports large-scale simulation and real-time scheduling of a high-fidelity operation model. And meanwhile, the entity is allowed to adopt different physical models or behavior models for simulation as required. The problem that in a traditional scheduling method, when a combined combat system adopts step length to advance logic time, the real-time performance is easily affected is solved.
2) The multi-core parallel simulation engine system supporting the combined operation can automatically divide a scheduling table according to the running period of a model and the system step length in a dynamic and static combined simulation scheduling method.
3) The multi-core parallel simulation engine system supporting the joint combat has the advantages that the independence of all nodes is strong, a central node for dynamically distributing loads does not exist generally, and simulation entities operated by all the nodes are well pre-deployed; the simulation engine is positioned on each node and used for scheduling the node.
4) The multi-core parallel simulation engine system supporting combined operation can directly input damage of an operation entity and the condition of generating a new entity (plane-launched missile) into a simulation scheduling method, adjust a scheduling table of a simulation engine and avoid the condition of excessive load.
5) The multi-core parallel simulation engine system supporting the joint operation can be applied to the scheduling of the joint operation simulation engine and can also be applied to the real-time scheduling of other background models.
6) The multi-core parallel simulation engine system supporting the joint operation improves the expandability of the original simulation platform by utilizing the HLA integrated interface, realizes network distributed simulation and heterogeneous simulation resource reuse based on the HLA standard, not only can meet the complexity requirement of the joint operation task, but also can improve the building and operating efficiency of the simulation system and promote the fusion and simulation response capability of the simulation resources.
7) The multi-core parallel simulation engine system supporting the combined operation supports the operation of the simulation system, can simulate the operation process and the force deduction and research and check the operation strategy, carries out the efficiency evaluation of the operation system according to the stored simulation data, and provides decision reference for the real and complex multi-arm combined operation.
8) The multi-core parallel simulation engine system supporting the combined operation mainly meets the requirements of the research and development of weapon equipment of various units and the exercise training of troops in China. The simulation system disclosed by the invention not only comprises the force model, but also realizes the functions of simulation scenario, simulation deployment, visualization, data acquisition, efficiency evaluation and the like.
Drawings
FIG. 1 is a schematic diagram of the description of the components of the present invention and the modular components of the simulation system;
FIG. 2 is a diagram of the model scheduling interaction of the user interface and the simulation engine of the present invention;
FIG. 3 is a schematic diagram of load balancing for a scheduling method employed in the present invention;
FIG. 4 is a representation of a dispatch plan generated by the present invention;
FIG. 5 is a simulation scheduling framework of the present invention;
FIG. 6 is a logical framework diagram of the emulation engine of the present invention;
FIG. 7 is a diagram of the core components of the emulation system of the present invention;
FIG. 8 is a solid model framework in the simulation system of the present invention;
FIG. 9 is a diagram of a combined combat model combat embodiment according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
In the following description, the present invention will be described in terms of various aspects, and a developer in the art may implement the present invention using a part or all of the structure or process according to the development need. The invention has been described in terms of construction frameworks, sequences, and so forth, for a clear understanding of the developer, but it will be apparent that the invention may be practiced without these specific details. In other instances, well-known features have not been described in detail so as not to obscure the invention.
The invention mainly researches the real-time scheduling problem of the multi-weapon combined combat model and the time propulsion problem which needs to be considered after the operation support environment is introduced. The logic time and simulation step size concepts of running the support environment impose constraints on scheduling. The invention provides a framework for separating time propulsion and a model scheduling process, and designs a dynamic and static combined scheduling algorithm based on step length load balancing, thereby obtaining good real-time performance and scheduling performance.
Aiming at the defects and shortcomings of the real-time scheduling method of the simulation engine system in the multi-weapon combined combat at present, the invention provides a multi-core parallel simulation engine system supporting modules, integration and multiplexing. The multi-weapon-species combined combat model has the following characteristics: firstly, the simulation entity is formed by assembling a plurality of models, such as physical models of ships, submarines, tanks, combat vehicles, antiaircraft guns, helicopters, airplanes and the like; meanwhile, the method comprises behavior models of advancing, firing, command control and the like; part of models are executed according to cycles, and the models generally use most of computing resources in a simulation engine system; setting an initial entity set, wherein the set can change in the simulation process; the invention adopts a load distribution mode combining dynamic and static states, namely, distribution according to needs. The invention combines the above 4-point characteristics of the battle model to obtain good super real-time effect and improve the dispatching operation efficiency.
The combat simulation engine system of the invention is divided from the aspect of function realization as shown in figure 1, and mainly comprises the following functional modules: the system comprises a scheduling model management module, a thread management module, an external interface management module, a high-level system structure management module, a visualization module, an evaluation module and a data recording module.
A scheduling model management module: the method is mainly responsible for analyzing the test model, loading a model dynamic library and the like, managing the model in a queue mode for the thread management module to call, and sending a signal to the thread management module when the model simulation settlement is completed. The scheduling model management module realizes simulation time management and simulation model promotion, responds to simulation user operation, and manages and schedules the simulation process. The simulation time management provides a time advance management function, so that the advance of real world time is not limited by simulation temporary or stop or simulation variable step length, and parallel advance of simulation entities is supported; while supporting time-step and event-step based propulsion. The combined combat simulation model mainly comprises a series of entity classes such as tanks, airplanes, radars, artillery, shell ships and warships, and each basic entity class corresponds to an entity management class which controls the creation, destruction, configuration, user interface and related data of the entities.
A thread management module: and the system is responsible for creating and scheduling threads, multithread scheduling takes out the model from the model queue for model simulation settlement after simulation starts, and obtains an operation signal of the external interface management module for scheduling the threads and the model. In order to reduce the mutual preemption problem among multiple threads, the phenomenon of dynamic load change can be adapted by a mode of actively sharing the load on the basis of a thread pool. Namely, the entity to be executed is actively acquired in real time, so that the resources are fully utilized, and the dynamic change of the entity resolving time is adapted.
External interface management module: and the scheduling model management module is used for realizing interaction between the user interface and model scheduling of the simulation engine system.
The scheduling model management module, the thread management module and the external interface management module form a scheduling module of the simulation engine system.
A high-level architecture management module: an HLA integrated interface is provided, external simulation interaction is carried out with the operation supporting environment, the expandability of the simulation platform is improved, and network distributed simulation and heterogeneous simulation resource reuse based on an HLA standard are realized.
A visualization module: the system is mainly used for dynamically representing a simulation process, can receive entity state change messages from a simulation engine so as to change the position or the state of a corresponding icon, and can send simulation data to a visualization computer through the Ethernet.
An evaluation module: the system is used for providing an analysis and evaluation interface for analysts, such as dynamically counting various warfare losses of all parties, evaluating possible future development trends and the like. It may receive various entity state changes from the simulation engine to dynamically update the statistics.
A data recording module: a blackboard-based interaction mode is employed for recording the operation of all model applications during system operation.
The multi-core parallel simulation engine system supporting the joint combat runs on the nodes arranged in the Ethernet distributed mode. The system runs on each node as shown in fig. 2, and human-computer interaction is realized through a master control platform. The computer where the master control platform is located serves as a node of the Ethernet, a multi-core parallel simulation mode is adopted in the node, the node serves as an HLA simulation federate member to operate in an operation supporting environment, the master control platform performs external simulation interaction with the operation supporting environment through an HLA integrated interface, and performs collaborative simulation with other member nodes under the control of HLA master control. The general control platform of a plurality of nodes can participate in the same high-level system structure simulation federation for collaborative simulation.
The simulation strategy adopted by the multi-core parallel simulation engine system comprises the following steps of one to three, and space and time resources are arranged for a simulation entity and a model component thereof.
The method comprises the following steps: and distributing the target nodes for the simulation entities, namely distributing the simulation entity models of all the nodes so that the total computation of the models on each node is equivalent.
The specific implementation comprises the following steps; 1) calculating the utilization rate of the processor of each simulation entity, and sequentially arranging the utilization rates from large to small; 2) initializing the idle rate of each node to be 1; 3) taking down the first simulation entity of the entity queue, distributing the first simulation entity to the node with the maximum idle rate, and updating the idle rate of the node; 4) and repeating the step 3) until the linked list of the entity queue to be distributed is empty. And if the idle rate of a certain node is less than 0, stopping allocating the additional simulation entity for the node.
Step two: the scheduling model management module generates a scheduling schedule of each node, namely, the model is allocated with simulation step length. So as to satisfy the requirement of large-scale simulation that a single node can contain hundreds of thousands of models, and simultaneously has higher processor utilization rate, and is particularly suitable for a large number of small models, which accords with the characteristics of large-scale simulation because the simulation scale will be degraded if the single model runs for a long time.
Step three: the scheduling table is adjusted by the scheduling model management module, namely the simulation step length of the destroyed entity and the generated new entity is adjusted, the global scheduling problem is converted into the local scheduling problem, the stability characteristic of the distributed and parallel simulation engines is guaranteed, and meanwhile the load is reduced to the maximum extent.
The simulation engine system runs on a high-level architecture, the real-time scheduling method is based on a load balancing principle, the load balancing principle is shown in figure 3, and scheduling is optimized through changing of simulation step length to reduce simulation running time. After model scheduling begins, model scheduling queues test models, n threads are generated for processing, models are taken from each thread for simulation, whether the models are empty or not is judged firstly, if yes, scheduling of the threads is finished, if not, model functions are called, output of the models is settled, a database is updated according to output results, the processes of taking the models and judging whether the models are empty or not are repeated, and when operation of taking the models from all the threads for simulation is finished, scheduling is finished.
The invention distributes simulation step length for the model based on the load balancing principle, and carries out static scheduling according to the prior information of the model, namely, a scheduling table is determined in advance before operation. The method for generating the scheduling schedule of the nodes by the scheduling model management module comprises the following steps 1-8. Set of periodic model formations { M } of all entities on a nodek(ii) a And k is 1,2, … and n, n is the total number of the models on the nodes, the models on the nodes share s different operation periods, s is less than or equal to n, and n and s are positive integers.
Step 1: and aligning the running periods of the models to realize efficient scheduling. The invention is a running period T of the modeliAdd one constraint, align by 2:
Ti＝2i·Tstep，i＝1,2,…,s (1)
wherein, TiThe model is the ith running period of the model, namely the step length of the model. T isstepThe system step size represents the time interval used when the simulation process advances in a step-size manner. E.g. system step size Tstep50ms, then the model step size TkCan be 50ms, 100ms, 200ms, 400ms, 800ms, etc.
Step 2: processor idle rate idle of each system step in initial scheduling period k1, k is more than or equal to 1 and less than or equal to TotalSteps, so as to ensure that the processor of each step has full processing capacity. Total Steps is a scheduling period TkThe total number of system steps involved, as follows:
TotalSteps＝Tk/Tstep (2)
and step 3: a scheduling schedule Table is generated. The scheduling plan Table is organized according to different model periods, including model step size and the number of system step sizes, i.e. sub-step sizes, included in the model step size. Using the linked list as a data structure, the scheduling schedule Table can be represented as:
Table＝{Mpq} (3)
wherein p is the system step size number in the model operation period, p is more than or equal to 1 and less than or equal to S, and S is the model operation period TkTotal number of system steps in; q is the element number, i.e. the model number, q is greater than or equal to 1 and less than or equal to LocalSteps [ i-]，LocalSteps[i]Is to store the step length T with the same modeliAnd the number of models executed within the same system step. Table element MpqIs shown in the model operation period TkAnd scheduling the model with the number q in the p-th system step.
As shown in fig. 4, model 1, model 2 and model 3 contain 2, 4 and 8 system steps, respectively. The Table element of the scheduling plan Table Table corresponding to the model step length of 50ms is M11And M12The Table element of the scheduling plan Table Table corresponding to the model step length of 100ms is M13、M15、M24And M26. Take the model step size of 100ms as an example, where M13And M15，M24And M26For two parallel system steps, model 3 and model 5 are scheduled in the first system step, and model 4 and model 6 are scheduled in the first system step.
The scheduling schedule Table is organized according to different model run times, and the following steps illustrate how to generate.
And 4, step 4: the various operation periods T of the model are sequentially arranged from small to large according to the number1,T2,…,TsTraversing in sequence; the following steps are carried out:
(4.1) calculating the number of step sizes contained in each operation period, i-th operation period TiThe number of system steps L involved is:
L＝Ti/Tstep,1≤i≤s； (4)
(4.2) creating each operating cycle TiScheduling Table of step length of each internal systemijJ is more than or equal to 1 and less than or equal to L. All schedule tablesijAnd forming a scheduling schedule Table. TableijThe model for the step size scheduling of each system is determined according to (4.3).
(4.3) will have a run period TiAll the models are sorted from large to small according to the execution time to form a queue, and the queue is traversed to execute the following steps;
(4.3.1) slave operation period TiStep size of each systemIdle rate idle of processor1,idle2,…idleLMaximum value of (1) is marked idlem；
(4.3.2) allocating the current model to the mth system step length, and recording the current model in TableimPerforming the following steps;
(4.3.3) setting the execution time of the current scheduled model as T, and updating the idle rate of the processor of the system step size and the subsequent affected system step size, namely
idlem+L*l-T/Tstep (5)
Wherein, idlem+L*lProcessor idle rate representing the m + L system step, 0 ≦ L<Ts/Ti。
And when the idle rate of the processor of the system step size is less than 0, the model cannot be scheduled, and the model needs to be stopped being distributed for the system step size and resources need to be increased.
The above allocation procedure follows a dual priority policy: models with short periodicity are higher in priority and models with longer average computation time are higher in priority. Each scheduling selects the sub-step with the highest idle rate to the model with the highest priority.
And 5: after the scheduling schedule is created, the simulation runtime can directly look up the table and execute the table, and assume that the current simulation time passes by Δ T ═ Ts1-Ts0For each operating cycle TiOf a schedule of calculating TiSystem step number s (i) to be currently executed:
S(i)＝[(Ts1-Ts0)modTi]/Tstep (6)
Ts1and Ts0Representing the simulation start time and end time of a certain model, mod is the remainder operation.
Step 6: setting each model MpqAt the moment of first run, the work can be completed in the initialization phase, and then the initial value P of the time of model runkComprises the following steps:
Pk＝-(q+1)Tstep (7)
let TkIs a certain moment of simulation time, model MpqSimulation time Δ T ═ Ts1-Ts0If the following condition is satisfied:
ΔT-Pk≥Tk (8)
executing the model, after the model is executed, PkIs set to Δ T.
And 7: when all T areiWhen the local load is balanced, all the operation periods reach the load balance.
Calculating TiDegree of unbalance θ ofiThe following were used:
wherein: l isijRepresents TiLoad of inner jth system step, i.e. TableijThe sum of the execution times of the assigned models;
The degree of imbalance μ for all operating cycles was calculated as:
wherein, PiIs the operating period TiThe longest execution time of the scheduled model.
For the processor utilization (load) of the model set, on a single processor, the requirement that the model set can meet the schedulability is mu < 1, on m identical processors, the requirement that the model set can meet the schedulability is mu < m, in engineering practice, the overhead of other functions of the system is also considered, which depends on various factors such as an operating system, operating system basic software, an operation support environment, a scheduling method and the like, so the maximum processor utilization cannot be achieved.
And 8: when the model set changes, the adjustment of the scheduling table is carried out.
Detecting before deleting entity if thetaiBeyond the imbalance threshold, the operating period T is reachediAnd migrating the model between the schedules in which the system step sizes with the highest internal load and the system step sizes with the lowest internal load are located. When an entity is newly built, after a node with the lowest load is selected, the system step length with the lowest load is selected, and a model is added to a corresponding scheduling table.
One basic issue with scheduling is selecting what granularity of parallelism: an entity or a model component of an entity. Because the execution of different entities is mutually independent tasks, the parallel operation is easy. And when the models are parallel, because a shared blackboard exists, performance loss is brought when concurrent protection is introduced, and in addition, the models are expected to be executed in a fixed sequence (such as task- > behavior- > physics) to reduce the phenomenon that the simulation result is uncertain caused by scheduling, so that a serial execution mode is adopted for the models in the same entity.
The basic structure of the engine scheduling function is shown in fig. 5, wherein the main components of the engine scheduling include:
1. and a Thread Pool (Thread Pool), determining the number of working threads through a configuration file during initialization, and creating a Thread object. Since the emulation engine also contains other service threads, the number of worker threads is typically the number of CPU cores minus 1.
2. The Main Thread (Main Thread) informs the Thread pool to start model solution when a step starts, and advances by one step after the end. And is also responsible for messaging through the emulation middleware.
3. Load Set (Workload Set), which maintains all the entities/units to be executed, which contain various model components: physical model, autonomic behavior (behavior model 1), task.
4. A worker Thread (Thread) (the Workload in the figure), acquires a unit or an entity from a Workload Queue model library, then executes the Tick function thereof, and finally traps the Tick of a specific model. After all the entity resolution is completed, the thread pool needs to inform the main thread backwards so that the main thread advances to the next step.
The logical framework of the simulation engine of the present invention is shown in FIG. 6, with modules with a gray background representing the user-implemented model, instantiated by the engine at initialization as desired. All executable modules inherit from the class of 'model', and provide standard interfaces such as initialization Init, step size scheduling Tick, event scheduling OnNotify and the like. The behavior Agent can use an "inference engine" to accomplish regular inference or action planning. Through the Ethernet dynamic simulation experiment system, the mutual cooperation process between protocols in Ethernet data communication can be dynamically reflected, and the data communication process can be better completed through the Ethernet. Data encapsulation is an important stage of protocol work, and the encapsulation process of data is dynamically and really reproduced in an experimental system.
The invention simultaneously supports the realization of a multi-weapon (sea, land and air) combined combat simulation system. The simulation can be carried out on the battle process and the battlefield situation. Based on the combat model provided by the invention, corresponding model development, simulation engines and a planning and editing tool are realized. As shown in fig. 7, the relationship between the various components of the system is given. The behavior of a combat entity involves a large number of elements, such as the role of the entity, the state of the entity, interactions between entities, tactical rules, etc., and how to correctly recognize these elements is a problem that must be solved before modeling can be performed. This link is called model development, which is located between the real world and the computer model, bridging the bridge between the domain user and the modeler. Atomic behavior, combinatorial behavior, state machine behavior are computer implementation level.
In a simulation system, the present invention uses a blackboard structure to enable communication between combat models. The blackboard provides a common workspace for managing data interactions between different federal members, a sharing system for combat missions and a result sharing system. As FIG. 8 illustrates a mockup framework, when an activity in a task is performed, an atomic behavior can send a trigger to the blackboard, informing the behavioral model of interest to begin working, and ultimately invoking the physical model indirectly by the behavioral model to produce the actual action. Since the functionality of the physical model can be accurately characterized from the outside, services can be provided using a standardized interface. The calling is indirectly realized through a simulation engine so as to ensure the transparency between models.
Each solid model contains a blackboard structure inside. The mockup can either be work driven by blackboard events (e.g., damage) or scheduled periodically (e.g., maneuver). Useful elements of the blackboard are:
1. and the Trigger (Trigger) represents the processing requirement from the task or the internal model, and is deleted after being used.
2. Facts (Fact), representing situational awareness data, persist and can be updated.
3. The command (Order), command data from the upper level, is deleted after use.
4. And the Message (Message) is used for the cooperation between the entities and is deleted after being used.
The blackboard system provides operations including:
1. order (Subscribe), ordering blackboard elements of interest.
2. Publish, Publish data that exists for a short time, such as triggers, messages.
3. Update (Update), Update fact.
4. Notification (Notify), when an event occurs on the blackboard, of the fighting model of interest.
An activity Agent can be formally represented as:
Agent＝<ID,Meta,Subscription,Publications,Data,Provided Cap,Required Cap>
wherein, ID is the unique mark of the model; the Meta comprises information such as applicable platform, compiling level, name and description; a Subscription is a set of blackboard element types of interest; publications are a set of published blackboard element types; data is the Data used by the model, which may be the rules of engagement for behavioral models, and performance parameters of equipment for physical models; the Provided Cap is the capability Provided by the Agent. For example: driving; required Cap is the capability Required by the Agent. For example: and (4) maneuvering.
Because the blackboard component adopts a similar interactive cooperation mode for the inside and the outside of the entity, the entity can also show the Agent characteristic. The interaction between entities is still in essence the interaction of Agent components. In this regard, the entities herein encompass the connotation of a traditional combat Agent model.
For a combat entity that emulates a runtime, it can be defined as:
Entity＝<ID,Type,Superior,Role,Capabilities,Tasks,Agents,Blackboard>
wherein, the ID is the unique identification of the entity in the scenario; type is an entity Type, usually expressed in the model number of the equipment; the method is generated by an entity assembly tool, and defines the agents required by the equipment; superior is a Superior identifier; role is a Role of an entity in an organization, such as a lead plane, a bureaucratic plane, a lead car; the Capabilities are a capability set possessed by the entity, are quantitatively expressed and can provide support for behavior decision; tasks are task lists that the entity needs to execute; agents is a set of behavior and physical model components contained in an entity and is determined according to Type; the Blackboard is a Blackboard structure shared by each Agent and the task model in the entity.
Similarly, the war Unit can be defined:
Unit＝<ID,Type,Superior,Subordinates,Role,Capabilities,Tasks,Agents,Blackboard>
the Type of the unit is generated by the unit assembly tool, which defines the Agent Type that makes up the unit. Subordinates contains the subordinate, which may be a subunit or an entity. Furthermore, for non-aggregation level cells, Agents only contain behavior Agent models. The unit itself corresponds to a virtual object that performs the command on behalf of a command entity (e.g., a long car). This has the advantage that when a long car is destroyed, the take over of the command role only requires the tagging and user logical processing of another entity (if required), without the need for model migration at the simulation system level.
Examples
As shown in fig. 9, the model structure of the aerial interception mission of the formation of an airplane is given, comprising three parts of a combat mission, a formation unit and airplane entities (valetudinarians and bureaucratins).
In air-to-air combat, a defense party (through a patrol plane or a ground radar) plans a route of a fighter according to information such as speed and direction of the fighter after detecting an attack attempt of the enemy, the route is generally calculated by taking the shortest encounter time as a target, then the fighter flies along the appointed route, and when the enemy is close to the fighter, an airborne radar needs to be turned on to start a search mode to prepare for access to the enemy. The mission itself can be divided into four activities, namely, air route planning, flying along the air route, radar operation and air-to-air fighting. Wherein the radar is powered on during flight, so that the two are required to be in parallel. The four activities in the interception task have a relatively fixed pattern and can be depicted by a flow chart. Assume that a campaign is currently being performed. The engagement itself has considerable uncertainty, and the specific process depends on the autonomous completion of the airplane entity.
When the battle activity is entered, the corresponding trigger is sent to the blackboard of the unit, and the command Agent receives the notification and carries out corresponding processing. It is assumed here that the "command Agent" is merely tagged to allow attacks when enemy opportunities are subsequently discovered. When an airplane entity flies in a battle area, the found enemy information is stored in a blackboard of the entity as an enemy fact by a sensor Agent (radar), and the enemy is reported to a formation unit in a specified format by a message Agent later. The information Agent of the unit fuses and evaluates the situation of the information reported by the two airplanes to form the situation fact which is stored in the blackboard; then the command Agent makes decision according to situation and tactical rule set, decides to adopt the fighting scheme of beyond visual range or near visual range and the motor tactics, and finally issues the strategy to the valentine plane and the bureaucratic plane respectively in the form of command.
After the plane blackboard receives the command of 'executing tactical maneuver', the plane blackboard triggers 'driving Agent' to call 'maneuvering Agent' to set direction, speed and the like according to maneuvering knowledge defined by the action set, so that the appointed tactical maneuver is completed. As the enemy changes continuously, once the firepower control Agent finds that the attack condition is met, the firepower control Agent calls the weapon Agent to shoot the weapon. If a "sensor Agent" (radar warning system) of the aircraft receives a threat from an enemy, such as a missile attack, the "threat" data is sent to the blackboard and triggers the "piloting Agent" to evade. The driving Agent has the multi-target priority ordering and conflict resolution capability. If a missile explodes, an explosion event is sent to the blackboard to trigger a 'damage Agent' to calculate the damage condition of the airplane.
In this example, it can be seen that the behaviour of the aircraft is not solidified, but is strained on a case-by-case basis. Each Agent module has independence, works together, and completes response and cooperation through the blackboard. If the sensor component is damaged, the entity automatically loses the detection capability only by deleting or stopping the sensor Agent component; further, the entity reaction behavior will be affected, and even the formation behavior will be affected.
Finally, the following description is provided: the above examples are only intended to illustrate the technical solution of the invention and not to limit the technical process, the invention can be extended and modified in application, therefore all modified and varied examples are within the spirit and teaching of the invention.
Claims (7)
1. A multi-core parallel simulation engine system supporting joint combat runs on nodes distributed by adopting Ethernet, and is characterized by comprising a scheduling model management module, a thread management module, an external interface management module and a high-level architecture management module; the scheduling model management module manages and schedules the simulation process of the model, manages the model in a queue mode for the thread management module to call, and sends a signal to the thread management module when the model simulation settlement is finished; the thread management module creates and dispatches threads, and multithread dispatching takes out the model from the model queue for simulation settlement; the external interface management module is an interface for interaction between the scheduling model management module and the outside; the high-level system structure management module provides a high-level system structure HLA integrated interface, performs external simulation interaction with an operation support environment, and realizes the collaborative simulation with member nodes;
the multi-core parallel simulation engine system firstly allocates target nodes for simulation entities, so that the total computation of models on each node is equivalent; then, a scheduling model management module generates a scheduling schedule of each node, a simulation step length is distributed to the model, the scheduling schedule is adjusted in the simulation process, and the simulation step lengths of the destroyed entity and the generated new entity are adjusted;
the scheduling model management module allocates simulation step length for the model, and the method comprises the following steps: aligning the running periods of the models; initializing the idle rate of a processor of each system step length in a scheduling period to be 1; generating a scheduling schedule, wherein the scheduling schedule is organized according to different model operation cycles and comprises a model step length and a system step length number contained in the model step length;
the scheduling model management module aligns the operation cycles of the models according to 2 and represents the following:
Ti＝2i·Tstep，i＝1,2,…,s
wherein, TiRepresents the i-th run cycle, T, of the model on the nodestepS represents that the model on the node uses s different operation periods;
the scheduling model management module also sets the time of the first operation of each model and an element M in a scheduling schedulepqRepresenting model with scheduling number q in the P-th system step in a certain operation period, setting initial value P of scheduled model operation timekComprises the following steps:
Pk＝-(q+1)Tstep
let TkIs a certain moment of simulation time, model MpqIs Δ T, if the following condition is satisfied:
ΔT-Pk≥Tk
executing the model, after the model is executed, PkIs set to Δ T;
the interior of each model comprises a blackboard structure; the communication between the models uses a blackboard structure;
one activity Agent is formally represented as:
(ID, Meta, description, Publications, Data, rendered Cap, Required Cap), wherein ID is the only mark of the model; the Meta comprises an applicable platform, an establishment level, a name and description information; a Subscription is a set of blackboard element types of interest; publications are a set of published blackboard element types; data is Data used by the model, fighting rules for the behavioral model and performance parameters of equipment for the physical model; the Provided Cap is the capability Provided by Agent; required Cap is the capability Required by the Agent;
because the blackboard part adopts a similar interactive cooperation mode for the inside and the outside of the entity, the entity also shows the Agent characteristic, and the interaction between the entities is still the interaction of the Agent components in essence; the Agent modules of all the entities have independence, the Agent modules of all the entities work together, response and cooperation are completed through the blackboard, and when a certain entity is damaged, the entity only needs to delete or stop the Agent module of the entity, and then the entity automatically loses the detection capability.
2. The simulation engine system of claim 1, wherein the allocating of the target node for the simulation entity is specifically: 1) calculating the utilization rate of the processor of each simulation entity, and sequentially arranging the utilization rates from large to small; 2) initializing the idle rate of each node to be 1; 3) taking a first simulation entity of an entity queue to be distributed, distributing the first simulation entity to a node with the maximum idle rate, and updating the idle rate of the node; 4) repeating 3) until the queue of the entity to be allocated is empty; and if the idle rate of a certain node is less than 0, stopping allocating and adding the simulation entity for the node.
3. The simulation engine system of claim 1, wherein the scheduling model management module generates the scheduling schedule for different operation periods T of the model in a descending order1,T2,…,TsTraversing, creating a scheduling table of each system step length in each operation period, and finally forming a scheduling schedule table;
for operation period TiCalculating the operating period TiThe number of system steps contained in L-Ti/TstepI ═ 1,2, …, s; will have a run period TiAll the models form a queue according to the sequence of the execution time from large to small, traverse the model queue, and execute the following (1) to (3) to create the operation period TiScheduling Table of step length of each internal systemij,1≤j≤L；
(1) Slave operation period TiFor each system step size, selects the maximum idle among the processor idlenessm；
(2) The current model is allocated to the mth system step length and recorded in the TableimPerforming the following steps;
(3) updating the processor idle rate of the mth system step size and the subsequent affected system step size; the processor idle rate update of the m + L system step is: idlem+L*l-T/Tstep(ii) a Wherein l is more than or equal to 0<Ts/Ti；
And when the idle rate of the processor of the system step size is less than 0, stopping distributing the model for the system step size.
4. The simulation engine system of claim 1, wherein the scheduling model management module directly performs table lookup for the running period T in the simulation runtime after the scheduling schedule is creatediThe system step number s (i) that should currently be executed is calculated as:
S(i)＝[(Ts1-Ts0)modTi]/Tstep
wherein, Ts1And Ts0Representing the simulation start time and end time of a certain model, mod is the remainder taking operation.
5. The simulation engine system of claim 1, wherein the scheduling model management module calculates the operating period T according to the following equationiDegree of unbalance θ ofi：
Wherein L isijRepresents TiThe load of the jth system step in the series,
and
maximum and minimum of the load for the jth system step, respectively;
the scheduling model management module calculates the degree of imbalance μ for all operating cycles according to the following equation:
wherein, PiIs the operating period TiThe longest execution time of the intra-scheduled model;
when the local load of all the operation periods is balanced, all the operation periods reach load balance.
6. The simulation engine system of claim 1, wherein the scheduling model management module adjusts the scheduling schedule, comprising:
detecting before destroying the entity, if the operation period T isiExceeds the imbalance threshold, during the operating period TiOn the corresponding scheduling table, migrating the model between the system step length with the highest load and the system step length with the lowest load;
when an entity is newly built, a node with the lowest load is selected, then a system step length with the lowest load is selected, and a model is added to a corresponding scheduling table.
7. The simulation engine system of claim 1, wherein the system further comprises a visualization module, an evaluation module, and a data logging module; the evaluation module is used for providing an analysis and evaluation interface for an analyst.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710417606.9A CN107193639B (en) | 2017-06-05 | 2017-06-05 | Multi-core parallel simulation engine system supporting combined combat |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710417606.9A CN107193639B (en) | 2017-06-05 | 2017-06-05 | Multi-core parallel simulation engine system supporting combined combat |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107193639A CN107193639A (en) | 2017-09-22 |
| CN107193639B true CN107193639B (en) | 2020-11-24 |
Family
ID=59877319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710417606.9A Active CN107193639B (en) | 2017-06-05 | 2017-06-05 | Multi-core parallel simulation engine system supporting combined combat |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107193639B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038628B (en) * | 2017-12-28 | 2023-02-03 | 华北计算技术研究所（中国电子科技集团公司第十五研究所） | Military exercise drill quantitative evaluation system |
| CN108335024B (en) * | 2018-01-23 | 2020-08-04 | 华中科技大学 | Shipboard aircraft continuous action task planning method and task planning description method |
| CN109656267B (en) * | 2018-12-24 | 2022-03-15 | 中国航空工业集团公司西安飞机设计研究所 | Parallel cooperative test method for flight control system |
| CN109800054B (en) * | 2018-12-24 | 2023-05-26 | 四川知周科技有限责任公司 | Distributed parallel real-time simulation scheduling realization method |
| CN110717263A (en) * | 2019-09-27 | 2020-01-21 | 中国人民解放军海军大连舰艇学院 | Combat model management system |
| CN111177892B (en) * | 2019-12-11 | 2023-05-02 | 中电普信(北京)科技发展有限公司 | Distributed simulation system |
| CN111400895B (en) * | 2020-03-12 | 2023-03-17 | 上海机电工程研究所 | Multi-level and multi-granularity cross-domain joint simulation event scheduling method and system |
| CN111522731B (en) * | 2020-03-13 | 2023-06-23 | 中国电子科技集团公司第二十九研究所 | Model integration method and device for online reloading of simulation model |
| CN111597035B (en) * | 2020-04-15 | 2024-03-19 | 北京仿真中心 | Simulation engine time propulsion method and system based on multithreading |
| CN111625935A (en) * | 2020-04-30 | 2020-09-04 | 苏州启明可视科技有限公司 | Chemical disaster accident real-time intervention simulation method and system |
| CN111858026B (en) * | 2020-06-10 | 2021-08-31 | 中国人民解放军海军航空大学航空作战勤务学院 | Efficient parallel scheduling method for large-scale multi-granularity simulation model |
| CN112131730B (en) * | 2020-09-14 | 2024-04-30 | 中国人民解放军军事科学院评估论证研究中心 | Fixed-grid analysis method and device for intelligent unmanned system of group |
| CN112270083B (en) * | 2020-10-23 | 2023-03-07 | 中国人民解放军海军航空大学 | Multi-resolution modeling and simulation method and system |
| CN112559153B (en) * | 2020-12-21 | 2024-05-28 | 北京仿真中心 | Multithreading and multiprocessing integrated simulation model component scheduling method and system |
| CN112925575A (en) * | 2020-12-29 | 2021-06-08 | 中国航空工业集团公司沈阳飞机设计研究所 | Airborne simulation operation method based on framework and plug-in |
| CN112560292A (en) * | 2021-01-06 | 2021-03-26 | 中国人民解放军63863部队 | Simulation system of complex environment target |
| CN113297729B (en) * | 2021-05-13 | 2022-08-09 | 中国人民解放军军事科学院战争研究院 | Parallel simulation entity partitioning method based on entity types |
| CN113781856B (en) * | 2021-07-19 | 2023-09-08 | 中国人民解放军国防科技大学 | Training simulation system for combined combat weapon equipment and implementation method thereof |
| CN114282366B (en) * | 2021-12-24 | 2022-11-22 | 中国人民解放军军事科学院战争研究院 | Aerial formation and entity two-stage resolution simulation modeling method |
| CN114757057B (en) * | 2022-06-14 | 2022-08-23 | 中国人民解放军国防科技大学 | Multithreading parallel combat simulation method and system based on hybrid propulsion |
| CN115964131B (en) * | 2023-03-16 | 2023-05-16 | 中国人民解放军国防科技大学 | Simulation model management system supporting multiple simulation engines and simulation model scheduling method |
| CN116108757B (en) * | 2023-04-04 | 2023-07-21 | 中国电子科技集团公司第十五研究所 | Multistage simulation time pushing method in training environment, server and storage medium |
| CN116562054B (en) * | 2023-07-06 | 2023-10-13 | 西安羚控电子科技有限公司 | Construction method and device of multi-entity collaborative real-time simulation system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101788919A (en) * | 2010-01-29 | 2010-07-28 | 中国科学技术大学苏州研究院 | Chip multi-core processor clock precision parallel simulation system and simulation method thereof |
| CN103593516A (en) * | 2013-10-30 | 2014-02-19 | 中国运载火箭技术研究院 | Combat system modeling and simulation system |
| CN104866374A (en) * | 2015-05-22 | 2015-08-26 | 北京华如科技股份有限公司 | Multi-task-based discrete event parallel simulation and time synchronization method |
| CN105630578A (en) * | 2015-12-24 | 2016-06-01 | 中国人民解放军海军航空工程学院 | Distributed multi-agent system-based combat simulation engine |
| CN106295084A (en) * | 2016-09-29 | 2017-01-04 | 北京华如科技股份有限公司 | Service-oriented expansible combination type artificial engine |
| CN106446427A (en) * | 2016-09-29 | 2017-02-22 | 北京华如科技股份有限公司 | Combined type battling entity modelwith control module as core and establishment method thereof |
-
2017
- 2017-06-05 CN CN201710417606.9A patent/CN107193639B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101788919A (en) * | 2010-01-29 | 2010-07-28 | 中国科学技术大学苏州研究院 | Chip multi-core processor clock precision parallel simulation system and simulation method thereof |
| CN103593516A (en) * | 2013-10-30 | 2014-02-19 | 中国运载火箭技术研究院 | Combat system modeling and simulation system |
| CN104866374A (en) * | 2015-05-22 | 2015-08-26 | 北京华如科技股份有限公司 | Multi-task-based discrete event parallel simulation and time synchronization method |
| CN105630578A (en) * | 2015-12-24 | 2016-06-01 | 中国人民解放军海军航空工程学院 | Distributed multi-agent system-based combat simulation engine |
| CN106295084A (en) * | 2016-09-29 | 2017-01-04 | 北京华如科技股份有限公司 | Service-oriented expansible combination type artificial engine |
| CN106446427A (en) * | 2016-09-29 | 2017-02-22 | 北京华如科技股份有限公司 | Combined type battling entity modelwith control module as core and establishment method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 《计算机生成兵力模型的实时调度技术》;吴雨淋 等;《北 京 航 空 航 天 大 学 学 报》;20140731;第41卷(第2期);246-251页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107193639A (en) | 2017-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107193639B (en) | Multi-core parallel simulation engine system supporting combined combat | |
| CN106446427B (en) | Accuse the combined type operation physical model and its construction method for core | |
| CN107967134B (en) | Novel combat effectiveness simulation modeling method | |
| Samad et al. | Software-enabled control: information technology for dynamical systems | |
| CN105630578A (en) | Distributed multi-agent system-based combat simulation engine | |
| Tidhar et al. | Flying together: Modelling air mission teams | |
| Cameron et al. | Rule-based peer-to-peer framework for decentralised real-time service oriented architectures | |
| Paunicka et al. | The OCP-an open middleware solution for embedded systems | |
| CN114169142A (en) | Task meta-model construction method, database and device for general combat process | |
| CN114326827B (en) | Unmanned aerial vehicle cluster multitasking dynamic allocation method and system | |
| CN114282833A (en) | Rules-based hierarchical task planning method for air-sea combined combat action | |
| Yun et al. | Formation and adjustment of manned/unmanned combat aerial vehicle cooperative engagement system | |
| CN113919068A (en) | Task-based aviation equipment support system simulation evaluation method | |
| CN112433806A (en) | Computing model structure and scheduling execution method for cloud simulation service | |
| CN106919386B (en) | The method and apparatus of code is generated based on ARINC653 operating system | |
| Jang et al. | An actor-based simulation for studying uav coordination | |
| Zheng et al. | Multiple task planning based on TS algorithm for multiple heterogeneous unmanned aerial vehicles | |
| CN114757057B (en) | Multithreading parallel combat simulation method and system based on hybrid propulsion | |
| Oh et al. | AddSIM: A new Korean engagement simulation environment using high resolution models | |
| CN116166225A (en) | Task planning middle stage frame design method based on software definition | |
| Li et al. | Software architecture of C2 mission system based on MAS and SOA for manned/unmanned aerial vehicle formation | |
| Ni et al. | An integrated architecture design method for multi-platform avionics system | |
| CN114077476B (en) | Multi-platform elastic avionics system cloud system and method | |
| Findler et al. | Distributed air-traffic control. II: Explorations in test bed | |
| Chenyan et al. | Research on multi-resolution modeling of command and control simulation system based on components |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |